Type 1 diabetes (T1D) develops from the autoimmune destruction of pancreatic beta cells by self-reactive T cells. Potentially self-reactive T cells escaping elimination in the thymus may subsequently be deleted or tolerized in the periphery. Recently, stromal cells of lymph nodes were reported to control self-tolerance by inducing the ectopic expression of various peripheral tissue antigens (PTAs) and presenting them to T cells in a manner that induces tolerance (deletion or regulation). Currently, both the transcriptional control of PTA expression in lymph nodes and the dysregulation of peripheral tolerance in T1D are unclear. Our laboratory recently identified Deaf1 as a transcriptional regulator of PTA gene expression in peripheral lymph nodes. Our findings show that the gene expression of PTAs, including insulin and other beta cell antigens, are down- regulated in the pancreatic lymph nodes (PLN) of non-obese diabetic (NOD) mice at 12 weeks of age. This change in expression coincides in time with the onset of beta cell destruction by autoreactive T cells. Preliminary studies show that Deaf1 regulates the expression of multiple PTA genes in the PLN. Thus far, two forms of Deaf1, a full-length functional form and a non-functional variant (Deaf1-VAR) have been identified in the PLN of 12 week-old NOD mice. At this age, Deaf1-VAR expression is significantly higher and Deaf1 expression is significantly lower in the PLN of NOD compared to NOD.B10 mice. Deaf1-VAR is expressed in the cytoplasm, has little transcriptional activity of its own, and inhibits the transcriptional activity of Deaf1 by heterodimerizing with it and retaining it in the cytoplasm. Remarkably, we identified an equivalent alternatively spliced Deaf1 isoform in the PLN of T1D patients. This human variant which was ~20-fold more abundant in the PLN of T1D patients than controls, behaves like the mouse Deaf1-VAR. The high expression of this variant also correlated with the absence of insulin gene expression in the PLN of T1D patients. Based on these data, it is hypothesized that Deaf1 controls the expression of and maintains peripheral tolerance to certain pancreatic beta-cell antigens, including insulin, in the PLN, and that a change in Deaf1 splicing contributes to the pathogenesis of T1D. To examine this hypothesis, proposed studies address the following: 1) What cell type in the PLN expresses Deaf1, Deaf1-VAR and PTAs? 2) How does Deaf1 mediate PTA gene transcription on a molecular level? 3) Does inflammation of the PLN or various "genetic factors" regulate the splicing of Deaf1? 4) Will knock-out of Deaf1 expression in PTA-expressing cells exacerbate NOD disease, and if so, is this due to a lack of autoreactive T cell deletion or a lack of regulatory T cell development? Finally, do either mouse models of autoimmune uveoretinitis or human APS 1 patients have changes in Deaf1 that can be correlated with disease? The results of these studies will provide an overall understanding of how Deaf1 and its isoforms maintain peripheral tolerance and thus, how a change in the splicing of Deaf1 or a mutation in DEAF1 may contribute to the development of autoimmune disease. PUBLIC HEALTH RELEVANCE: Type 1 diabetes can result from a break in peripheral tolerance that is, in part, mediated by the expression of peripheral tissue antigens (PTAs) on lymph node cells. We have identified a transcriptional regulator, Deaf1, that regulates the expression of PTAs in the pancreatic lymph node (PLN), and have data to suggest that the alternative splicing of this gene leads to reduced PTA expression in the non-obese diabetic mouse model and in human type 1 diabetic patients. The studies proposed here will examine the role of Deaf1 in maintaining peripheral tolerance and help us understand how a change in Deaf1 splicing may contribute to the pathogenesis of T1D.